We will use std::unordered_map from the standard template library to implement our symbol table

We separate the parser into a generic FileReader class and a set of C-style translation functions

The assembler class performs two passes through the assembly code. Although this makes the algorithm O(3N) instead of O(2N) (because we separate the first pass from the book into two steps: one to read the file and another to build the symbol table), the assembler class is completely self-contained. It can thus be used to convert assembly instructions to machine code without the need of an intermediate file.

Implement a generic file reader that you will re-use in your next assignments

Implement functions for translating assembly instructions to machine code

change_extension(const char* filename, const char* newext, char* outfilename, int maxsize)

bool open(const char* filename): Opens the file associated with filename for reading. The associated file pointer is a member variable (mFp). If the file cannot be opened, set mFp to NULL and set mHasMoreLines = false. If the file is successfully opened, this method should call advance(). This function returns true if the file was successfully opened and false otherwise. Lastly, if a file is opened when this method is called, the current file should be closed before opening a new one.

~FileReader(): closes the file corresponding to mFp

bool hasMoreLines() const: returns true if we have not reached the end of the file yet

const char* line() const: returns a constant pointer to the last line read (stored in mLine)

void advance(): Calls fgets to get the next line and stores the result in mLine. The return value of fgets should be used to update mHasMoreLines. This function should also strip comments and whitespace from the line. See the class notes for details.

InstructionType cmd_type(const char* line): returns either A_INSTRUCTION, C_INSTRUCTION, or L_INSTRUCTION depending on the contents of line

void cmd_symbol(const char* line, char* value, int size): extracts the symbol from line into the string value. size is the max number of characters that can be stored in value. The behavior of this function is undefined if line does not contain either an L_INSTRUCTION or A_INSTRUCTION.

void cmd_dest(const char* line, char* value, int size): extracts the destination from line into value. The behavior is undefined if line is not a C_INSTRUCTION.

void cmd_comp(const char* line, char* value, int size): extracts the computation from line into value. The behavior is undefined if line is not a C_INSTRUCTION.

void cmd_jump(const char* line, char* value, int size): extracts the jump from line into value. The behavior is undefined if line is not a C_INSTRUCTION.

void write_jmp_code(const char* jmp, uchar code[16]): writes the jump bits to code based on the given string (e.g. "JMP", "JGE", etc)

void write_dest_code(const char* dest, uchar code[16]): writes the destination bits to code based on the given destination string (e.g. "D", "A", etc)

void write_comp_code(const char* comp, uchar code[16]): writes the computation bits to code based on the given computation string (e.g. "M+1", etc)

void write_a(int value, uchar code[16]): writes an A instruction corresponding to the passed decimal value.

initSymbolTable should initialize the symbol table with Hack’s built-in symbols. (See your book or our previous lab for details).

translate(instructions) should compute the type and machine codes for each instruction using two passes

write(const char* filename, instructions) should write the machine code to a file with the given filename.